The Implication of Repeated Versus

Continuous Strain on Nerve

Function in a Rat Forelimb Model
Mitsunobu Watanabe, MD, Makio Yamaga, MD, Teiji Kato, MD,
Junji Ide, MD, Toshio Kitamura, MD, Katsumasa Takagi, MD,
Kumamoto, Japan

We studied the effect of repeated and continuous nerve strain using a rat forelimb model to
investigate whether an innocuous level of strain applied continuously affects nerve function
when applied repeatedly. We used the rat medial cord of the brachial plexus and assessed the
effects of strain by studying nerve histology (blood–nerve barrier), function (grasping strength),
and electrophysiology. Continuous stretching was applied to the rat forelimb for 1 hour at 2
N. After this strain neither histologic analysis, grasping strength, nor electrophysiologic
analysis revealed any effect. We then applied repeated strain at both 60 and 120 times per
hour; after the latter strain abnormalities in histology, grasping strength, and nerve conduction
were identified. There results suggest that a small nerve strain applied repeatedly results in
nerve dysfunction. Our data may help explain the cause of nonspecific neural symptoms in the
upper extremities of patients with no objective findings. (J Hand Surg 2001;26A:663– 669.
Copyright © 2001 by the American Society for Surgery of the Hand.)
Key words: Nerve implication, repeated strain, brachial plexus, rat.

Patients who complain of neurogenic symptoms
such as pain and numbness of the neck, shoulder, and
upper limbs are often encountered in daily clinical
practice. With the recent progress in diagnostic tech-
niques, the number of cases leading to confirmed
diagnoses has increased but there are still many cases
in which a definite cause cannot be established. Pa-
tients with frequent use of upper limbs because at
occupation or sports activities manifest such a vari-
ety of symptoms. For lack of a better term, patients
From the Department of Orthopaedic Surgery, School of Medicine,
Kumamoto University, Kumamoto, Japan.
Received for publication March 1, 2000; accepted in revised form
December 29, 2000.
No benefits in any form have been received or will be received from
a commercial party related directly or indirectly to the subject of this
article.
Reprint requests: Mitsunobu Watanabe, MD, Department of Ortho-
paedic Surgery, School of Medicine, Kumamoto University, 1-1-1
Honjo, Kumamoto, 860-8556 Japan.
Copyright © 2001 by the American Society for Surgery of the Hand
0363-5023/01/26A04-0008$35.00/0
doi:10.1053/jhsu.2001.24142
with nonspecific neurogenic symptoms in the ab-
sence of objective findings may be characterized as
having an overuse syndrome. The pathogenesis,
however, remains obscure.1 Because of the difficulty
in understanding the cause of symptoms in such
disorders, we have developed an experimental model
to evaluate the effect of continuous and repeated
nerve strain.
Materials and Methods
We conducted a preliminary study to determine
the adequate nerve strain that causes no nerve disor-
der when applied alone. Thereafter we applied con-
tinuous or repeated nerve strain as determined in the
preliminary study. We evaluated nerve function in
the rat forelimb medial cord (Fig. 1).
Animal Preparation
The animals were handled in accordance with the
National Institutes of Health Guide for Care and Use
of Laboratory Animals and the local institution

The Journal of Hand Surgery 663

664 Watanabe et al / Repeated Nerve Strain Injury

Figure 1. (A) The whole rat forelimb. The arrow indicates nerves of the upper arm. (B) A close-up view of the nerves in
rat upper arm. MC, medial cord of the brachial plexus; MA, medial anterior thoracic nerve; MN, median nerve; UN, ulnar
nerve. The arrow indicates the elbow joint.

guidelines for humane use of animals in research. A
total of 142 male Wistar rats weighing 300 to 350 g
were used. Anesthesia was induced by intraperito-
neal administration of pentobarbital sodium (40
mg/kg body weight) under conditions of natural res-
piration. With the rats supine on a traction device
designed by our group the medial cord of the brachial
plexus was stretched indirectly with a steel wire, 0.35
mm in diameter, inserted into the radius and ulnar
distally. When pulled in the direction of 80° abduc-
tion of the shoulder, the brachial plexus is elongated
linearly.2 We then evaluated the effects of continu-
ous or repeated strain on the brachial plexus.
The repeated strain was applied at 2 levels of
frequency: 60 and 120 cycles per hour (Fig. 2).
Continuous strain for 1 hour was used as the control.
The traction was performed through a steel wire
inserted in the forearm; it was measured by the actual
weight applied to the steel wire. The acceleration and
deceleration of the traction load were 10 N/min for
both the continuous and repeated strain in all exper-
iments.
Preliminary Study
We determined the traction force necessary to
cause a mild nerve strain when applied singly for
 The Journal of Hand Surgery / Vol. 26A No. 4 July 2001 665
Figure 2. The methods of traction.
1 hour. No nerve disorder should be observed with
this strain after traction is released. One-hour con-
tinuous traction was performed at traction forces
of 1, 2, 5, and 0 N (control group). Twenty-four
rats were used for this study (6 rats for each
group). Compound nerve action potential (CNAP)
of the brachial plexus was measured at 10-minute
intervals during traction and over 30 minutes after
release of traction. The amplitude did not change
significantly during traction in the 1 N traction
group (Fig. 3). In the 2 N group the amplitude
Figure 3. Changes in CNAP amplitude at traction forces
of 0, 1, 2, and 5 N in 60 minutes of continuous traction in
the medial cord of brachial plexus of the rat. Data are
presented as mean values ⫾ SD. *p ⬍ .01, **p ⬍ .05
versus the 0 N group; 1-way ANOVA followed by Scheffé
post hoc test.
decreased to 60% of the basal level but recovered
rapidly after release of the traction load. In the 5 N
group the amplitude did not recover to the original
level even by 30 minutes after release of the trac-
tion. Based on these results 2 N was defined as the
traction force in this study.
Continuous or Repeated Strain on Forelimb
Nerve
The rat forelimb was pulled at 2 N continuously or
repeatedly. We evaluated nerve function in the rat
forelimb medial cord by histologic analysis (blood–
nerve barrier), functional testing (grasping strength),
and electrophysiologic analysis. Repetitive stretch-
ing at the strain identified in the preliminary study
included both 60 and 120 cycles per hour with no
interval between the stretching.
Percentage Elongation of the Nerve
The nerve elongation rates for the group given
repeated stretching 120 times and group given con-
tinuous traction were measured according to the
method of Lundborg and Rydevik3 using 20 rats (10
rats/group). Statistical analysis was performed by the
Mann-Whitney U test.
Figure 4. Percentage elongation of the rat brachial plexus
in the group given continuous streching and the group with
repeated traction applied 120 times (n ⫽ 10/group). Data
are presented as mean values ⫾ SD.
666 Watanabe et al / Repeated Nerve Strain Injury
Figure 5. (A) A medial cord of the brachial plexus before stretching. (B) Nerves are stretched by a 2-N load. The arrows
indicate the points of fine suture marking.
Histologic Study
In histologic study blood–nerve barrier (BNB) per-
meability was assessed with Evans blue albumin (EBA)
using 24 rats. The rats were divided to continuous
stretching group, 60 times repeated stretching group,
and 120 times repeated stretching group (8 rats/group).
The BNB permeability was assessed using serum albu-
min mixed with Evans blue. The 1% EBA was pre-
pared by mixing 5% bovine albumin (Sigma Chemical
Co, St Louis, MO) with Evans blue (Merck, Darmstadt,
Germany). After removing free Evans blue through a
Sephadex column (Pharmacia Biotech, Uppsala, Swe-
den) before use, 1 mL/body weight 100 g was injected
intravenously from the femoral vein. After perfusion
for 1 hour the rat was killed, and the medial cord of the
brachial plexus and the median nerve were dissected.
 The Journal of Hand Surgery / Vol. 26A No. 4 July 2001 667

Figure 6. Fluorescence microscopic appearance. (A) A longitudinal section of a normal nerve. The EBA is confined to the
lumen of the endoneurial capillary (arrows). (B) A medial cord of the brachial plexus in the group stretched 120 times. The
BNB is deteriorated and EBA is diffusely spread into the endoneurial space. The arrow indicates the endoneurial capillary.

The dissected tissues were fixed in 5% formalin for 24
hours. After freezing the tissues 10-␮m sections were
cut and mounted in 50% glycerin in water and observed
under a fluorescent microscope using a dual-band mir-
ror unit (Provis, Olympus, Tokyo, Japan). The EBA
leakage into the endoneurium was evaluated in accor-
dance with the method of Rydevik and Lundborg.4
Functional Study
Forelimb nerve function was assessed by the
grasping strength test. Fifty rats were used for this
study. Five groups were prepared: continuous trac-
tion, 60 or 120 times repeated traction, no traction
and no wire inserted (control), and no traction and
wire inserted (sham operation) (10 rats/group).
Twenty-four hours after release of traction a grasping
strength test was performed similar to the method of
Bertelli and Mira5 to assess the rat forelimb nerve
function. The rats were lifted by the tail and allowed
to grasp a steel wire grid connected to an electronic
balance by 4-digit grasping with the wrist and elbow
joint extended. The absolute value on the balance
scale was recorded the moment the grip was loos-
ened. A 1-way ANOVA followed by the Scheffe
post hoc test was used for statistical analysis.
Electrophysiologic Study
Approaching the rat upper arm medially we ex-
posed the medial cord of the brachial plexus with
care so as not to damage any vessels or connective
tissues around the nerve. A bipolar electrode (KU-
97087; Unique Medical Inc, Tokyo, Japan) for stim-
ulation was placed on the proximal side of the medial
anterior thoracic nerve junction of the medial cord.
668 Watanabe et al / Repeated Nerve Strain Injury

elongation rate did not differ with the traction

method. Figure 5 shows a nerve elongated by 2-N
traction.

Histologic Study
The BNB function was normal in the continuous
stretching group and showed no extravascular leak-
age of EBA. In the group stretched 60 times leakage
of EBA occurred in 4 of 8 rats (50%). In the group
stretched 120 times breakdown of the BNB function
was noted in 7 of 8 rats (88%) (Fig. 6). Thus, BNB
function was maintained during continuous stretch-
ing but impaired with repeated stretching.

Figure 7. Grasping strength of the rat forepaw. Data are
presented as mean values ⫾ SD. *p ⬍ .01, 1-way ANOVA
followed by Scheffé post hoc test.
Functional Study
Neither continuous nor repeated (60 times/h)
stretching resulted in significant differences in grasp-
ing strength; however, when nerve strain was re-
peated 120 times per hour grasping strength de-
creased to 68% compared with the continuous strain
group (Fig. 7).
Electrophysiologic Study
Compound nerve action potential studies showed
no significant decrease of amplitude in the continu-

In addition, a recording electrode was placed 5 mm

proximal to the stimulating electrode. Compound
nerve action potentials were recorded with a Neu-
ropack Four Mini (Nihon Kohden, Tokyo, Japan)
using a supramaximal stimulus; the local nerve tem-
perature was maintained at 37°C with a heat lamp.
Compound nerve action potentials were recorded
before traction and immediately after release of trac-
tion and are expressed as a percentage of the basal
level before traction. The repeated stretching was
performed at frequencies of 60 or 120 times per hour.
Continuous stretching for 1 hour was used as the
control. Twenty-four rats were used for this study (8
rats/group). Numerical values obtained are presented
as the means ⫾ SD. A 1-way ANOVA followed by
Scheffe post hoc test was used for statistical analysis.
p values of ⬍ .05 were considered significant.

Results
Percentage Elongation of the Nerve
The nerve elongation rate showed no difference
between the continuous traction (8.6%; range, 4.8%
to 10.3%) and the traction repeated 120 times (8.5%;
range, 5.0% to 10.2%) (Fig. 4). Thus, the nerve
Figure 8. Changes in CNAP amplitude in the brachial
plexus after traction. Data are presented as mean values ⫾
SD. *p ⬍ 0.01, 1-way ANOVA followed by Scheffé post
hoc test.
 The Journal of Hand Surgery / Vol. 26A No. 4 July 2001 669
ous stretching group. In the group stretched 60 times
amplitude decreased significantly to 75% of the basal
level. In the group stretched 120 times the decrease
in amplitude was more pronounced (46% of basal
level) (Fig. 8). Thus, CNAP amplitude was signifi-
cantly lower under repeated stretching and the de-
crease was more marked in the group that underwent
120 repetitions.
Discussion
Two newtons is considered to be the amount of
load that produces no nerve disorder after traction
when applied for 60 minutes. In the 60-minute con-
tinuous stretching group at 2 N the nerve conduction
disorder returned to normal immediately after release
of the traction load. No significant changes in neither
grasping strength nor BNB function were observed.
Even such mild stretching forces, however, provoked
a breakdown of the BNB function and decreased the
grasping strength and amplitude of nerve conduction
when they were applied repeatedly for 120 times per
hour, although light microscopy showed normal me-
dial cords of the brachial plexus. Furthermore, the
nerve disorder becomes more pronounced as the
frequency of repetition increases. Possibly, the nerve
disorder develops when the mechanical stress is ac-
cumulated by repeated loading to exceed a certain
threshold with an increase in the frequency of
stretching.
The elongation rate of 8% in the brachial plexus
represents the lower stretch limit because the venous
flow in the nerve begins to be disturbed at this point.
Therefore, 8% elongation is regarded as critical for
the peripheral nerve.3 This 8% rate, however, may
only be safe when stretch is applied continuously for
1 hour, because repeated strain during the same time
interval led to a breakdown in BNB function. This
breakdown causes an increase in the internal pressure
of the nerve and abnormal ion equilibrium, eventu-
ally resulting in nerve dysfunction.6 –10
These results suggest that the rat brachial plexus is
vulnerable to repeated stretching. Although the anat-
omy of rats differs from human anatomy, the bra-
chial plexus of rats is also located near the shoulder
joint, which has the largest range of motion, and the
surrounding connective tissue is sparse and easily
elongated.11
Our study may provide a basis to explain the
obscure pathogenesis of nonspecific upper-extremity
neurogenic symptoms in humans. Pain and numb-
ness around the neck, shoulder, and upper extremity
may originate from repetitive strain experienced by
the brachial plexus during activities of daily living or
work exposure. Our model suggests that a continu-
ous strain for 1 hour may allow neural recovery, but
the same strain applied repeatedly results in greater
neural dysfunction. Our data do not show that a more
sustained continued strain would be innocuous but
suggest that low level of strain repeated in a given
time period may have greater implication on neuro-
dysfunction.
The authors thank Takayoshi Maekawa, Nobuki Murayama, and
Miki Ohgushi for advice, suggestions, and comments.
References
1. Quintner J, Elvey R. The neurogenic hypothesis of RSI. In:
Bammer G, ed. Discussion papers on the pathology of
work-related neck and upper limb disorders and the impli-
cations for diagnosis and treatment. National Centre for
Epidemiology and Population Health working paper no.
24. Canberra, Australia: National Centre for Epidemiology
and Population Health, The Australian National Univer-
sity, 1991:1–25.
2. Kitamura T, Takagi K, Yamaga M, Morisawa K. Brachial
plexus stretching injuries: microcirculation of the brachial
plexus. J Shoulder Elbow Surg 1995;4:118 –123.
3. Lundborg G, Rydevik B. Effects of stretching the tibial
nerve of the rabbit: a preliminary study of the intraneural
circulation and the barrier function of the perineurium.
J Bone Joint Surg 1973;55A:390 – 401.
4. Rydevik B, Lundborg G. Permeability of intraneural mi-
crovessels and perineurium following acute, graded exper-
imental nerve compression. Scand J Plast Reconstr Surg
1977;11:179 –187.
5. Bertelli JA, Mira JC. The grasping test: a simple behavioral
method for objective quantitative assessment of peripheral
nerve regeneration in the rat. J Neurosci Methods 1995;
58:151–155.
6. Lundborg G. Intraneural microcirculation. Orthop Clin
North Am 1988;19:1–12.
7. Lundborg G. Nerve injury and repair. Edinburgh: Churchill
Livingstone, 1988:32–101.
8. Mizisin AP, Kalichman MW, Myers RR, Powell HC. Role
of the blood-nerve barrier in experimental nerve edema.
Toxicol Pathol 1990;18:170 –185.
9. Myers RR, Heckman HM, Powell HC. Endoneurial fluid
is hypertonic. J Neuropathol Exp Neurol 1983;42:217–
224.
10. Rechthand E, Rapoport SI. Regulation of the microenvi-
ronment of peripheral nerve: role of the blood–nerve bar-
rier. Prog Neurobiol 1987;28:303–343.
11. Kitamura T, Takagi K, Yamaga M, et al. Experimental
study on the interneural circulatory changes of the brachial
plexus after a stretch. Shoulder Joint 1994;18:1– 4.